API Mabilis na Pagsisimula — curl, Python, TypeScript

Ang pinakamabilis na paraan upang masuri ang paggana ng API — curl:

Karaniwang Kahilingan:

curl https://gate.joingonka.ai/v1/chat/completions \\n  -H "Content-Type: application/json" \\n  -H "Authorization: Bearer jg-iyong-key" \\n  -d '{\n    "model": "Qwen/Qwen3-235B-A22B-Instruct-2507-FP8",\n    "messages": [\n      {"role": "user", "content": "Ano ang Gonka?"}\n    ]\n  }'

Streaming (tugon nang paunti-unti):

curl https://gate.joingonka.ai/v1/chat/completions \\n  -H "Content-Type: application/json" \\n  -H "Authorization: Bearer jg-iyong-key" \\n  -d '{\n    "model": "Qwen/Qwen3-235B-A22B-Instruct-2507-FP8",\n    "messages": [\n      {"role": "user", "content": "Sumulat ng hello world sa Python"}\n    ],\n    "stream": true\n  }'

Ang tugon ay dumarating sa format na JSON (karaniwan) o Server-Sent Events (streaming) — ganap na compatible sa OpenAI API.

Ang opisyal na OpenAI Python SDK ay gumagana sa JoinGonka Gateway nang walang pagbabago:

pip install openai

Karaniwang Kahilingan:

from openai import OpenAI\n\nclient = OpenAI(\n    base_url="https://gate.joingonka.ai/v1",\n    api_key="jg-iyong-key",\n)\n\nresponse = client.chat.completions.create(\n    model="Qwen/Qwen3-235B-A22B-Instruct-2507-FP8",\n    messages=[\n        {"role": "user", "content": "Ipaliwanag ang blockchain sa simpleng salita"}\n    ],\n)\n\nprint(response.choices[0].message.content)

Streaming:

stream = client.chat.completions.create(\n    model="Qwen/Qwen3-235B-A22B-Instruct-2507-FP8",\n    messages=[{"role": "user", "content": "Sumulat ng sorting algorithm sa Python"}],\n    stream=True,\n)\n\nfor chunk in stream:\n    if chunk.choices[0].delta.content:\n        print(chunk.choices[0].delta.content, end="")

Tool calling:

import json\n\ntools = [{\n    "type": "function",\n    "function": {\n        "name": "get_weather",\n        "description": "Kumuha ng panahon sa lungsod",\n        "parameters": {\n            "type": "object",\n            "properties": {\n                "city": {"type": "string", "description": "Pangalan ng lungsod"}\n            },\n            "required": ["city"]\n        }\n    }\n}]\n\nresponse = client.chat.completions.create(\n    model="Qwen/Qwen3-235B-A22B-Instruct-2507-FP8",\n    messages=[{"role": "user", "content": "Anong panahon sa Moscow?"}],\n    tools=tools,\n)\n\ntool_call = response.choices[0].message.tool_calls[0]\nprint(f"Function: {tool_call.function.name}")\nprint(f"Arguments: {tool_call.function.arguments}")

Sinusuportahan ng Qwen3-235B ang native tool calling — tama ang pagtawag ng mga function, nang walang pag-parse ng mga sagot na text.

Pag-install:

npm install openai

Karaniwang Kahilingan:

import OpenAI from 'openai';\n\nconst client = new OpenAI({\n  baseURL: 'https://gate.joingonka.ai/v1',\n  apiKey: 'jg-iyong-key',\n});\n\nasync function main() {\n  const response = await client.chat.completions.create({\n    model: 'Qwen/Qwen3-235B-A22B-Instruct-2507-FP8',\n    messages: [\n      { role: 'user', content: 'Sumulat ng Express.js server' },\n    ],\n  });\n\n  console.log(response.choices[0].message.content);\n}\n\nmain();

Streaming:

const stream = await client.chat.completions.create({\n  model: 'Qwen/Qwen3-235B-A22B-Instruct-2507-FP8',\n  messages: [{ role: 'user', content: 'Ipaliwanag ang async/await' }],\n  stream: true,\n});\n\nfor await (const chunk of stream) {\n  const content = chunk.choices[0]?.delta?.content || '';\n  process.stdout.write(content);\n}

Tool calling:

const response = await client.chat.completions.create({\n  model: 'Qwen/Qwen3-235B-A22B-Instruct-2507-FP8',\n  messages: [{ role: 'user', content: 'I-convert ang 100 USD sa EUR' }],\n  tools: [{\n    type: 'function',\n    function: {\n      name: 'convert_currency',\n      description: 'Pag-convert ng mga pera',\n      parameters: {\n        type: 'object',\n        properties: {\n          amount: { type: 'number' },\n          from: { type: 'string' },\n          to: { type: 'string' },\n        },\n        required: ['amount', 'from', 'to'],\n      },\n    },\n  }],\n});\n\nconst toolCall = response.choices[0].message.tool_calls?.[0];\nconsole.log(`Function: ${toolCall?.function.name}`);\nconsole.log(`Arguments: ${toolCall?.function.arguments}`);

Lahat ng halimbawa ay gumagamit ng opisyal na OpenAI SDK — walang karagdagang libraries na kailangan. Palitan lang ang base_url at api_key.

Sinusuportahan ng JoinGonka Gateway ang lahat ng karaniwang parameter ng OpenAI Chat Completions API:

Mga parameter ng modelo ng Qwen3-235B: contextual window — 128K token, maximum na sagot — 2048 token. Buong detalye: HuggingFace. Ang listahan ng mga modelo ay magagamit sa pamamagitan ng GET /v1/models.

Dalawang endpoint:

• OpenAI-format: POST https://gate.joingonka.ai/v1/chat/completions
• Anthropic-format: POST https://gate.joingonka.ai/v1/messages

Authentication: Authorization: Bearer jg-iyong-key (OpenAI) o x-api-key: jg-iyong-key (Anthropic)

Ang format ng tugon ay ganap na katugma sa OpenAI at Anthropic — anumang SDK, library o framework na sumusuporta sa OpenAI o Anthropic API, ay gumagana sa JoinGonka Gateway nang walang pagbabago. Ang Claude Code ay kumokonekta sa pamamagitan ng Anthropic-format nang direkta.